About Me

YuXuan Wu・Horikita Saku National University of Singapore

A student aspiring to leverage AI for advancing natural and fundamental sciences.

Self Introduction

Autonym: YuXuan Wu; Pseudonym: Horikita Saku

Majored in artificial intelligence as an undergraduate.

Interested in the intersection of data science and natural sciences, particularly in the realms of biological information and astrophysical.

Currently conducting research on discrete representation and AI applications in bioinformatics and medicine at the National University of Singapore, including GWAS and single-cell analysis.

Kaggle Expert.

My personality

I appreciate quiet spaces, reading, listening to the rain and relishing a good cup of coffee.

My favorite book is If On a Winter’s Night a Traveler.

Probably a bit workaholic.

Publications

CodeUnlearn: Amortized Zero-Shot Machine Unlearning in Language Models Using Discrete Concept

YuXuan Wu, Bonaventure F. P. Dossou, Dianbo Liu

Under Review at ICLR · 2024

Pages

IceCube - Neutrinos in Deep Ice

The top 3% of all participating teams globally. First Silver Medal. My story in the competition.

2023-10-26

10 min read

Astronomy

Neutrinos

First time with the astronomical telescope.

Try to calibrate astronomical telescopes and make observations of Jupiter and Saturn.

2023-10-11

1 min read

Astronomy

Star Observation

Engineering Innovation competition -- 1 place in Shanghai

A rare team competition, an interesting experience.

2023-10-20

2 min read

CV

Image Segmentation

Experience

Kaggle Expert

Kaggle * 2

2023 - Present

After a year of active engagement within the Kaggle community and consistent contributions to code, achieved the distinction of Kaggle Notebook Expert.
Subsequently, earned two silver medals and progressed to the role of Competition Expert.

Visiting Scholarship

Cognitve AI for Science Team · National University of Singapore

2024.1-Present

investigating the application of Nonlinear Methods in Genome-wide association studies (GWAS).
VQ / Machine unlearning

IceCube - Neutrinos in Deep Ice - Silver Medal(top3%)

IceCube Neutrino Observatory

2023

Reconstruct the direction of neutrinos from the Universe to the South Pole

Achieved the 21th in the Neutrinos and Astrophysics competition, ranking in the top 3% globally among all participating teams.
Utilized a 3D point cloud convolution model based on the EdgeConv operator, developed various RNN models, and employed a multi-stage training method grounded in IceCube’s physical principles.
This achievement also marks my first medal in Kaggle competitions.

HMS - Harmful Brain Activity Classification - Silver Medal(top2%)

Harvard Medical School

2024

Developed a model trained on Electroencephalography (EEG) signals and Spectrogram recorded from critically ill hospital patients to classify a variety of harmful brain activities.

Achieved 38th place in the Neuroscience and Physiology competition, ranking in the top 2% globally among all participating teams.
Developed a 1D model based on EEG signals and employed innovative training methods to create an effective 1D+2D multi-modal model.

Quick Links